LLaVA-VSD: Large Language-and-Vision Assistant for Visual Spatial Description

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jin, Yizhang, Li, Jian, Zhang, Jiangning, Hu, Jianlong, Gan, Zhenye, Tan, Xin, Liu, Yong, Wang, Yabiao, Wang, Chengjie, Ma, Lizhuang
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912095023398912
author Jin, Yizhang
Li, Jian
Zhang, Jiangning
Hu, Jianlong
Gan, Zhenye
Tan, Xin
Liu, Yong
Wang, Yabiao
Wang, Chengjie
Ma, Lizhuang
author_facet Jin, Yizhang
Li, Jian
Zhang, Jiangning
Hu, Jianlong
Gan, Zhenye
Tan, Xin
Liu, Yong
Wang, Yabiao
Wang, Chengjie
Ma, Lizhuang
contents Visual Spatial Description (VSD) aims to generate texts that describe the spatial relationships between objects within images. Traditional visual spatial relationship classification (VSRC) methods typically output the spatial relationship between two objects in an image, often neglecting world knowledge and lacking general language capabilities. In this paper, we propose a Large Language-and-Vision Assistant for Visual Spatial Description, named LLaVA-VSD, which is designed for the classification, description, and open-ended description of visual spatial relationships. Specifically, the model first constructs a VSD instruction-following dataset using given figure-caption pairs for the three tasks. It then employs LoRA to fine-tune a Large Language and Vision Assistant for VSD, which has 13 billion parameters and supports high-resolution images. Finally, a large language model (Qwen-2) is used to refine the generated sentences, enhancing their diversity and accuracy. LLaVA-VSD demonstrates excellent multimodal conversational capabilities and can follow open-ended instructions to assist with inquiries about object relationships in images.
format Preprint
id arxiv_https___arxiv_org_abs_2408_04957
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LLaVA-VSD: Large Language-and-Vision Assistant for Visual Spatial Description
Jin, Yizhang
Li, Jian
Zhang, Jiangning
Hu, Jianlong
Gan, Zhenye
Tan, Xin
Liu, Yong
Wang, Yabiao
Wang, Chengjie
Ma, Lizhuang
Computer Vision and Pattern Recognition
Artificial Intelligence
Visual Spatial Description (VSD) aims to generate texts that describe the spatial relationships between objects within images. Traditional visual spatial relationship classification (VSRC) methods typically output the spatial relationship between two objects in an image, often neglecting world knowledge and lacking general language capabilities. In this paper, we propose a Large Language-and-Vision Assistant for Visual Spatial Description, named LLaVA-VSD, which is designed for the classification, description, and open-ended description of visual spatial relationships. Specifically, the model first constructs a VSD instruction-following dataset using given figure-caption pairs for the three tasks. It then employs LoRA to fine-tune a Large Language and Vision Assistant for VSD, which has 13 billion parameters and supports high-resolution images. Finally, a large language model (Qwen-2) is used to refine the generated sentences, enhancing their diversity and accuracy. LLaVA-VSD demonstrates excellent multimodal conversational capabilities and can follow open-ended instructions to assist with inquiries about object relationships in images.
title LLaVA-VSD: Large Language-and-Vision Assistant for Visual Spatial Description
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2408.04957